We propose to continue the follow-up of the Nurses' Health Study II (NHSII), a Cancer Epidemiology Cohort of 116,430 women enrolled at ages 25 to 42 years in 1989. The NHSII is a unique cohort in many ways, as the only large cohort of women enrolled before menopause and with nearly 30 years of repeated measures of exposures. We have information on smoking, weight, medication use, diet, physical activity and medical diagnoses updated every two to four years. Further, we have collected detailed data on in utero and infant exposures (from participants' mothers), diet during high school, and other unique exposure information. We have nearly complete ascertainment of deaths using the National Death Index and other methods. The follow- up of the NHSII cohort remains high, with 94% cumulative follow-up. The NHSII biorepository includes plasma, red blood cells, white blood cells, and first morning urine samples from 29,611 women; two thirds of these women contributed timed follicular and luteal premenopausal samples. In 2008 to 2011, 16,510 women provided second blood and urine samples when the majority were postmenopausal. We also have cheek cell DNA from an additional 29,859 women. Adding to these resources, we were recently funded to collect fecal samples from 25,000 participants for microbiome analysis. We have collected archival tissue blocks for incident breast and ovarian cancers and melanoma (approximately 70% of requests have been obtained), and collections for other cancers are ongoing. Breast and ovarian cancer tissues have been utilized for gene expression, and tissue microarrays are used extensively for immunohistochemistry. Prediagnostic mammograms have been collected for breast cancer cases and controls, as well as pathology material from previous biopsies of benign lesions. Substantial proportions of the cohort have GWAS and metabolomic data from nested case-control analyses. This cohort has been highly productive; over 330 papers have been published during the last five years, and the rate of publication is increasingly rapidly. With our active resource sharing component, we have participated in 19 consortia as a member of the NCI Cohort Consortium, one of the few cohorts contributing substantial data on premenopausal women and the largest contributor of premenopausal breast cancers. We also have active data sharing collaborations with individual external investigators. These external collaborations and consortia have resulted in more than 63 publications during the current funding period. In the current proposal, we aim to continue cohort follow-up and maintain these unique resources. We will continue collecting blood and urine samples from women with breast cancer who also provided prediagnostic samples, tumor tissue for incident cancers, and pre-diagnostic pre- and postmenopausal mammograms. We propose to conduct a long-term (2-3 year) reproducibility study for gut microbiome. Finally, we will enhance our data management and analytic tools to maximize efficiency and use of our complex resources to identify modifiable determinants of cancer incidence and survival.